摘要
针对传统仿生控制方法在轨迹控制方面存在效率差和易陷入局部最优等问题,设计了一种基于模糊PID的电力巡检机器人移动轨迹纠偏方法。先构建机器人的空间运动模型,通过机器人的当期位置定位和空间位姿确定轨迹偏差情况;将模糊控制理论与PID控制方法结构,优选关键参数及提升控制精度。为进一步精确机器人的轨迹误差,将传统的固定论域转换为可变论域提高模糊PID模型的适应度,最后引入了BP神经网络模型参与模糊推理和反模糊化,进一步改善机器人的轨迹控制精度。实验结果显示,模糊PID控制方法的距离和耗时分别为183 m和35 s显著优于传统的控制方法,同时提出算法下的机器人移动路径轨迹方差更接近于0值线。
Aiming at the problems of poor efficiency and local optimality of traditional bionic control methods in trajectory control,a fuzzy PID based correction method for moving trajectory of power inspection robots was designed.Firstly,the spatial motion model of the robot is constructed,and the trajectory deviation is determined by the current position positioning and spatial pose of the robot.The fuzzy control theory and PID control method are combined to optimize the key parameters and improve the control accuracy.In order to further accurate the trajectory error of the robot,the traditional fixed discourse domain is transformed into a variable discourse domain to improve the fitness of the fuzzy PID model.Finally,the BP neural network model is introduced to participate in fuzzy reasoning and defuzzification,so as to further improve the trajectory control accuracy of the robot.The experimental results show that the distance and time of the fuzzy PID control method are 183m and 35s respectively,which are significantly better than the traditional control method,and the trajectory variance under the algorithm is closer to 0 value.
作者
王健
WANG Jian(Datang International Power Generation Co.,LTD.Beijing Gaojing Thermal Power Branch,Beijing 100041,China)
出处
《自动化与仪器仪表》
2024年第12期209-213,共5页
Automation & Instrumentation
作者简介
王健(1976-),男,北京人,本科,总工程师,主要研究方向为智能控制,网络安全、数字化建设、安全监督管理等。
